【AI-for-Medical-Diagnosis】

★Project1:Chest X-Ray Medical Diagnosis with Deep Learning

1. Pre-process and prepare a real-world X-ray dataset
   
2. Use transfer learning to retrain a DenseNet model for X-ray image classification
   
3. Use a technique to handle class imbalance
   
4. Measure diagnostic performance by computing the AUC (Area Under the Curve) for the ROC (Receiver Operating Characteristic) curve
   
5. Visualize model activity using GradCAMs
   

★Project2:Evaluation of Diagnostic Models

1. Metrics
   -True Positives, False Positives, True Negatives, and False Negatives
   -Accuracy
   -Prevalence
   -Sensitivity and Specificity
   -PPV and NPV
   -ROC Curve
   
2. Confidence Intervals
   
3. Precision-Recall Curve
   
4. F1 Score
   
5. Calibration
   

★Project3:Brain Tumor Auto-Segmentation for Magnetic Resonance Imaging (MRI)

1. What is in an MR image
   
2. Standard data preparation techniques for MRI datasets
   
3. Metrics and loss functions for segmentation
   
4. Visualizing and evaluating segmentation models